SOME STUDIES ON AUTOMATIC GENERATION CONTROL OF MULTI-AREA INTERCONNECTED RESTRUCTURED POWER SYSTEMS

Abstract

The prime objective of automatic generation control (AGC) is to adjust the active power generation in response to variable power demands and hence to maintain scheduled system frequency and scheduled tie-line power flows with neighboring control areas at desired tolerance values. A sizeable fall in frequency might badly affect the timing of electric clocks, magnetizing currents in transformers/induction motors, constant speed of AC motors, continuous operation of processes and synchronous operation of various units in power system. Additionally, power system may face a serious instability problem at substantial drop in the frequency. In steady state, automatically these variations must be zero. Enhanced power system stability is achieved with the proper design of supplementary controller adopted in an AGC system. However, continuous growth in size and complexity, stochastically changing power demands, system modeling errors, alterations in electric power system structures and variations in the system parameters over the time has turned AGC task into a challenging one. Consequently, conventional control strategies may be incompetent to handle such unpredictable variations in an AGC system. Hence, the researchers over the world are trying to propose several novel control strategies that fuse knowledge, techniques and methodologies from varied sources to tackle AGC problem of power system effectively. The literature survey indicates that several researchers, to tackle AGC issue in restructured system, have presented various types of controllers optimized using various conventional and intelligent soft computing techniques. The literature survey also unveils that the performance of AGC system depends chiefly on the sort of intelligent technique exploited and structure of the controller. Hence, the goal of the present study is to propose different types of new vi

supplementary controller structures for various types of restructured as well as traditional power systems. The presented work is divided into ten chapters. Chapter 1 deals with the introduction of AGC topic in deregulated environment. Chapter 2 presents a critical review of AGC schemes in restructured power system. Chapter 3 stresses on the modeling of traditional and restructured power systems under the study. The main simulation work starts from Chapter 4. In Chapter 4, the study is firstly conducted on a proposed restructured two-area multi-source hydrothermal and hydrothermal gas power systems interconnected via AC and AC/DC parallel tie-lines. Modern optimal control theory based optimal PI structured controllers are designed with full state vector feedback control strategy employing performance index minimization criterion. From the results obtained in the study, it is substantiated that the use of AC/DC parallel links as an area interconnection shows enrichment in the dynamic performance of the system in terms of less oscillations, settling time and peak overshoots/undershoots in the deviation in frequency and tie-line power responses. Eigenvalue study confirmed the positive effect of AC/DC parallel links on the system dynamic performance and stability. It is also observed that the multi-source hydrothermal system shows inferior performance in comparison to the single-source thermal system due the presence of hydro source in each area of the multi-source hydrothermal system due to the non-minimum phase characteristics of hydro turbines. The full state feedback optimal PI controllers work well and are very much robust but in realistic environments, the measurement of all states is not feasible all the time. Hence, next, in Chapter 5, some modern methods are adopted to conduct the study. In first attempt, a modified fuzzy PI (FPI) controller optimized using genetic algorithm vii

(GA) is proposed for different electric power system models such as traditional two area non-reheat thermal, reheat thermal, multi-source hydrothermal and restructured two-area reheat thermal systems. In traditional two-area multi-source hydrothermal system, each control area owns two generating units, one non-reheat thermal and one mechanical governor based hydro power plant. However, in restructured two-area single-source system, each control area owns two single reheat thermal generating units. Firstly, a FPI-1 controller is designed with nominal range of membership functions (mfs) and GA tuned output scaling factors. Secondly, to test the impact of alteration in horizontal range of mfs of FPI-1, it is further optimized to get FPI-2 controller. The results of FPI-1 and 2 controllers are compared and the results due to later controller are found to be superior. Yet, FPI controllers are designed only for a traditional two-area non-reheat thermal system; they are successfully applied on other system under studies. The performance of FPI controllers is found significantly superior in terms of lesser numerical values of settling times (STs), peak undershoots (PUs) and various performance indices (PIs) compared to conventional controllers based on optimal, GA, gravitational search algorithm (GSA), bacterial foraging optimization algorithm (BFOA), hybrid BFOA-particle swarm optimization (hBFOA PSO) and hybrid firefly algorithm-pattern search (hFA-PS) techniques. Next, in Chapter 6, BFOA optimized fuzzy PI (FPI) and fuzzy PID (FPID) controllers are proposed for traditional two-area non-reheat thermal, reheat thermal, multi-source hydrothermal and restructured multi-source hydrothermal power systems. BFOA is used to simultaneously tune the input and output scaling factors of FPI/FPID controller keeping mfs and fuzzy rules invariant. It is observed that FPI controller shows superior results in terms of lesser values of STs/PUs/PIs compared to PI controller based on recently reported techniques like GA/PSO/BFOA/hBFOA viii

PSO/hFA-PS/FA/artificial bee colony (ABC) and FPI controller tuned using PS/PSO algorithms for the same system design. Further, a fractional order PID (FOPID) structured controller is suggested for AGC problem solution of power systems in Chapter 7. The parameters of FOPID controller are optimized exploiting BFOA. At first, a traditional two-area multi-source hydrothermal system is considered and the advantage of FOPID is established over PI/PID controller optimized using hFA-PS and PID controller optimized using grey wolf optimization (GWO) techniques. To show the effectiveness of the method, the approach is further extended to restructured two-area multi-source hydrothermal and thermal gas systems. The analysis of the simulation results discloses the efficacy of FOPID controller over BFOA/differential evolution (DE)/GA optimized PID controller. Then, the study is extended to a restructured three-area multi-source hydrothermal power system. In the next step of the study, a maiden attempt is made to propose a fractional order fuzzy PID (FOFPID) controller for traditional two-area multi-source hydrothermal, restructured two-area multi-source hydrothermal, restructured two-area multi-source thermal gas and restructured three-area multi-source hydrothermal AGC systems in Chapter 8. The parameters of FOFPID controller are also tuned utilizing BFOA. The critical analysis of the obtained results revealed the worth of FOFPID controller over FOPID controller in terms of less numerical value of STs, PUs and PIs. It is also experienced that FOFPID controller satisfies the AGC requirements in different power transactions taking place under deregulated environment more fruitfully than FOPID controller. In Chapter 9, FOFPID controller is implemented in AGC of restructured three area multi-source hydrothermal system considering appropriate generation rate ix

constraint (GRC), deadzone (DZ), boiler dynamics (BD) and time delay (TD). However, controller is optimized for linear system it works robustly in the presence of GRC/DZ/BD/TD physical constraints; though in the presence of GRC/DZ/BD/TD the system performance degraded drastically in comparison to the linear or the system with GRC only. Further, investigations clearly reveal that the controller is found to perform well when the system is subjected to higher degree of uncontracted load demands and simultaneous occurrence of uncontracted load demands. Thus, controller parameters obtained for the linear system are robust enough and need not be retuned for the system having appropriate GRC or GRC/DZ/BD/TD or wide changes in the size and location of contract violations. Thus, BFOA tuned FOFPID controller and other controllers proposed in the previous chapters may be options to supply reliable power with quality to the consumers. Finally, Chapter 10 presents an overview of the contributions made in the current thesis. Few suggestions are also given to extend the research in the future.